This repository provides a petroleum-engineer-friendly coverage map of the public Equinor Volve dataset.
It does not crawl the dataset. It uses a frozen metadata catalog and produces structured data-availability tables.
The Volve dataset is a field archive release, not a curated research dataset.
Before starting any study, you should know:
- Which wells actually have drilling telemetry?
- Which wells have structured daily drilling reports?
- Which wells have wireline logs?
- Which wells have reservoir models or seismic data?
This repo answers those questions.
Open:
outputs/coverage_matrix_binary.csv
Rows are wells. Columns are petroleum data categories. Value is 1 (present) or 0 (not present).
Use this to:
- Select wells for drilling studies (look for DDR_XML, WITSML or XML, and WELL_TECH or DOCS).
- Select wells for log-based studies (look for DLIS and/or LAS).
- Avoid data-poor wells.
Open:
outputs/coverage_by_well.csv
Columns:
- well
- total_files
- data_categories_present
Use this to identify high-coverage wells and avoid wells with narrow scope.
Open:
outputs/coverage_by_tag.csv
This tells you how the dataset is distributed across disciplines.
Examples:
- High WITSML count suggests drilling telemetry is abundant.
- DLIS and LAS indicate wireline logs exist.
- SEGY indicates seismic is present.
- System-level archive folders were excluded to prevent distortion of coverage statistics.
- Well identifiers were extracted from file paths.
- Field-level data (models, seismic packages, general documents) may not appear in well-level tables.
You may cite this coverage audit methodology as:
“Prior to analysis, a filesystem coverage audit of the Volve dataset was performed to identify wells with sufficient operational and logging data.”
See:
docs/dataset_coverage_audit.md